|
Wi-Fi-enabled laptops are ubiquitous. According to a January 2007 ChangeWave study of over 4000 cell phone users (Link to study).
Research in Motion (RIM) and Apple are "the primary beneficiaries of a seismic shift towards more advanced phones." Apple's iPhone is now the top choice for 17% of respondents planning to buy a new cell phone in the next six months. RIM is in second place with 15%.
Also, 72% of respondents reported that they are very satisfied with their iPhone, RIM is again a strong second with 55% saying that they are very satisfied. Also recently, Research In Motion announced that it added 15-20 percent more subscribers than it originally predicted for its fourth fiscal quarter.
Both iPhone and Blackberry have Wi-Fi interfaces. This propels Wi-Fi onto center stage for ubiquitous connectivity and means that Wi-Fi networks at the 'edge' are proliferating both in numbers and scale.
Why mesh?
802.11 Wireless Mesh Networks (WMNs) technology is for the broadband enterprise, multi-dwelling units, hotels, community and neighborhood networks.
It is also gaining significant attention as a possible way for Internet Service Providers (ISPs), carriers, and others to roll out robust and reliable wireless broadband service access in a way that needs minimal up-front investments. Properly implemented mesh has the capability to self-organize and self-configure.
Wireless Mesh Networks can be deployed incrementally, one node at a time, as the business case dictates. Wireless Mesh Networks consist of a set of access point (AP) nodes with the capability of receiving, routing and transmitting to provide end-to-end routes to every other node in the network through multiple hops. Mesh networks must maintain low overhead multi-hop routes.
Due to mesh's multi-hop nature, mesh APs cooperatively forward each other's packets, thus every packet has to be replicated multiple times through multiple nodes before reaching its destination. Total available throughput for a node is limited by the raw channel capacity and by the forwarding load (backhaul hops) imposed by other nodes.
Further, since wireless medium is a shared resource, the mesh exhibits contention delays since only one node can grant access to the wireless medium at a time. Data is thus relayed from one node to another, repeated in a store-and-forward manner.
Since two sufficiently distant radios can transmit concurrently, the total amount of data that can be simultaneously transmitted over one hop increases linearly with the total area of network.
The capacity thru single-radio mesh is somewhere between 1/N and  , where N is the number of hops on the way to a gateway node with a wired Internet connection. One hop means that there is one mesh forwarding instance.
Subsequent analysis assumes perfect mesh forwarding, no interference and perfect coordination of the Wi-Fi channel access, real world capacity will likely be lower. Assuming worse case , note how rapidly capacity decays from the number of mesh backhaul hops (see Figure 1).
Click here for Figure 1.
Figure 1: Peak Available Mesh AP Throughput as a Function of Number of Hops: One hop includes a single mesh forwarding instance.
There are four ways to minimize the number of hops:
- Drop in more wired backhaul—but that defeats the purpose of the mesh. Why mesh if wired backhaul is available?
- Add more radios to each node— this raises system cost and eliminates one additional hops' worth of throughput hit per each additional radio
- Migrate mesh node-to-node backhaul to 802.11n
- Maximize the distance between nodes with smart antennas. Smart antennas permit widening the inter-node spacing in a mesh by increasing the available signal strength at the receiver and mitigate interference by containing the signal towards each client and each node.
We'll next investigate in more detail and quantify the benefits of moving to 802.11n for mesh backhaul and adding smart antennas to mesh nodes.
|
